I ripped a cd of mine recently (EAC), and the last track had a few digipops that I wasnt too pleased with. I marked the times, and then I re-ripped, once on each of my drives, and they were still there. I cleaned the cd, still there. I listened to the cd, using both drives, and they were on the cd itself. Now, EAC ripped the track securly with %99.9 or %100 acuracy when I ripped it, so if it were a problem, per se, EAC would have picked it up as such and not given me a perfect or near perfect rip, right? I downloaded the problem track off a P2P and it DIDNT have those pops, though...

Now, aside from a little dust that i could clean off, the cd seems to be in tip top shape, no visible scratches on the audio part of the silver side. So, my question, if EAC reports a flawless or near flawless copy, could the cd still have errors?

Are you using a poorly implemented C2 (or none at all)?

Edit: added missing word 'using'.

I cant say im too sure; the drive that has C2 (MATSHITADVD-ROM SR-8584A) didnt find any errors on the disc when i scanned it, the rip comes out at worst %99.9 when i do a cd image, and the same goes for the song when i extract it by itself.

This isnt the first time this occured, too. I have another cd like it, though admittedly that other cd is a little more worn. But still, EAC gets it at %99.9, and yet there are pops.

I could say that if my cds arent damaged by ware, then i just have two faulty discs... but that seems so highly unlikely, seeing that the digipops are similar by nature on each disc.





Yes, theres no question that I'm not a poorly implimented C2

I think he meant that you should try using EAC with C2 error correction DISABLED so that EAC has to read everything twice and does not rely on potentially wrong C2 error information from your drive. Alternatively, try Burst mode + Test & Copy, if the checksums are identical, the flaw is in the disc and not your drive.

yeah, w/o the C2 there are still pops, and the test+copy has the same hash. oh well, guess its time to replace those cds

maybe im slow about this, but its still bothering me.... when EAC is reading the track, it reads the sections with the clicks perfectly; the error correction bar doesnt move at all and it keeps going like nothings wrong. this implies that the the information that gets written to the subsequent wave is a PERFECT duplicate of the information on the CD, correct? Now, if I hear the pops on the cd when i play it on 3 seperate cdrom drives, and i hear the pops when i extract wavs from all three of those cdroms with pretty much flawless rips... does it mean that

a> the pops were present when I bought the cd

or

b> the pops represent a flaw, such as a scratch or something, that happened after I bought it, that is so defined that its unremovable.

but if its b, shouldnt that cause a sync error, suspicious possition, or at ver least an error correction? or... am I missing something?

Look at the recorded data in a wave editor, it might offer some insight into what the pop is. You could also try accuraterip, if you are lucky that CD will be in the database (there are close to 1 million tracks now in the database) and you can tell if other cds have the same pop.

It's a); Otherwise there would be read and/or checksum errors

thanks for the accuraterip suggestion. It helped with one of the now 3 problem discs. Its too bad that im looking for verification, otherwise i couldve sent more information their way. Oh well.

my first cd with a track with a pop in it ripped 100%, and accuraterip gave it an accurate rating with a confidence 7. So that pop must be on all cds (Opeth - Deliverance - 06 - By The Pain I See In Others - 10:07 right channel), apparently.

the other cds though ripped fine (100% and 99.9%, hashes identical with a test+copy), but accuraterip gave it an inaccurate rating confidence 1, "Your disc may be a different pressing".

one of the two problem discs, as i said, I downloaded the track and there were no pops, so I know its not a problem with everyone's disc... Im just confused as to why I get a 99.9% or 100% on them if it IS a problem, you know? Part of me wonders why EAC isnt picking anything up yet im still hearing pops no matter where i listen to the album.



thanks, thats what I think too. what im going to do is ask on the band's forum. but still, i think im going to have to replace some of the discs.... doesnt make any sense to me.

Slightly less than 100% track quality in EAC can happen without any problem caused by the disc. I think it is a bug.

You don't know the source of that record. It might be from a single CD (different master ?) The lossy codec of the donwloaded file may hide clicks, too.
Replacing discs because of audible problems which are already on the CDs (your case, obviously) won't work. Most likely every (physically) different CD which you'll find will have the same problem.
When people are confronted with this problem for the first time, it's very hard to realize that 90% of all CDs sold for a lot of money have extremely shitty quality (e.g. such pops).

thanks... its just kind of hard for me to accept that a store bought cd would have flaws like that, but it is the industry we're talking about...

>as i said, I downloaded the track and there were no pops

Depends on where that track came from, if the artist was included on a compilation cd then it could have been mastered differently, perhaps that track is on more than one album? or a greatest hits?

The Gathering - If_Then_Else - 11 - Pathfinder isnt on any comps that i know of, nor on any other of their cds in that form. But I could have gotten a bad pressing, for all I know. Accuraterip claimed inaccurate, confidence 1, "may be different pressing", despite 99.9% and 100% track rips. so it makes sense that there may be multiple pressings out (the disc is from '00).

Same goes for Hotel Costes - Quatre (at least two tracks pop). Theres a regular pressing and a limited pressing, I got the limited one, so it it could be a matter of pressings. But to make 100% sure, I made a copy for my friend over a year ago, if he still has it im going to listen to his copy to see if his clicks as well. If it does, then i assume bad alt. pressing as well.

since youre an accuraterip person, question for you. if EAC rips a track 100%, yet accuraterip says inaccurate confidence 7, do I assume a> I have an alt pressing, b> accurate rip settings are off, or c> EAC ripped 100% despite error? If its an alt pressing, Id sure love to know how to differentiate, ie accuraterip storing a Catalgue# or an ISRC.

>if EAC rips a track 100%, yet accuraterip says inaccurate confidence 7,

From your previous message you said it was ripped accurate with a confidence of 7, is this theoretical?

>do I assume a> I have an alt pressing,

If the other tracks are also all off then AccurateRip will suggest it is a different pressing.

>b> accurate rip settings are off

Not if 7 others had that confidence

>c> EAC ripped 100% despite error?

It is possible for EAC to re-read the same data (with error) and report 100% even though there is an error.

>accuraterip storing a Catalgue# or an ISRC.

AFAIK these numbers identify a recording piece, a new number is not issued if that piece is re-mastered, or even if the piece is sold to a different company the ISRC stays the same.

My apologies, theoretical. I was playing with accuraterip on some other discs as well, ones with no audible problems.



... which swings me right back to the start of this thread, when i asked this question. So then, its possible to have a 100% rip on EAC when there are errors? So then how can I trust EAC at all? Accurate rip is a nice concept, but the cds that i wish to rip are either not part of the database or have too little referance for it to be of full use (innacurate, confidence 1 means nothing, since the other user could have had a ruined disc with a faulty extraction).

If this is true, then am I to assume that you cant rip a track 100% and actualy be sure its an "exact audio copy"?

EAC relies on the drive giving back different results on an error re-read and this is probabbly the case for majority of errors, but is it hard to say how many errors sneak under the radar (I think I saw a thread here about plextools pro even missing errors: plextor write the firmware and the ripping software, so if they don't have a chance what chance does eac have when it is paired with just any old CD ripping drive).

If you wanted accurate ripping with a cd not in accuraterip and have multiple drives (from different manufacturers) then it is fairly safe to say a different cd drive shouldn't read an error the same as another cd drive, so comparing crcs is your best bet.

I took your suggestion, and a funny thing happened... all 3 drives read the cd exactly the same, all tracks with the same CRC and same audio MD5s (shntool), with the exception of the last track, which was the problem track. But heres the funny thing... i ripped the last track with all 3 drives, and split it to 30 second pieces to compare shntool md5s of each piece, and with the exception of the last second of audio, they all matched perfectly. Its funny because the last second of music doesnt have any pops! in other words, all three drives ripped the section with the pops perfectly and exactly the same, but the track itself is damaged from something else at another point. Go figure!!

but doing this type of test on multiple drives is good, it tells me when ive got the best rip that i can get with my cd, regardless of comparisons with other people (accuraterip).

thanks for the help, all folks. I'll use this as a guide for my rips from now on.

Perfect 100% security is never possible, nowhere in science (a bit philosophy...). You can only increase the likelyhood of a perfect rip. You do this by trying to get the same results from different methods of ripping - the more different the methods, the better.
If you have only one drive and accurip isn't available or not confident enough, it's a good idea to rip in burst mode with test & copy - it is extremely unlikely that CRCs will match, unless the rip is perfect.
And don't worry about EAC's non-secure error correction - those cases are extremely rare.

That is an offset/overreading issue for sure. Has nothing to do with the CD but with your drives and EAC.Accurip is in general the best method to achieve a high probability of a perfect rip, because it allows comparison with physically different CDs - CDs that may not have the scratch that yours has.

The actual maths can be astounding, CRCs in accuraterip are 32 bit: so a roughly 1:4 billion possibilities on the CRC. The real probability increaser comes from confidences, for example:

8 people have the same cd and rip the cd with crcs of:

12345678
12345678
12345678
12345678
12345611
12345678
12345678
12345678

notice the guy in position 5, his cd had a scratch, but his results will not appear in the database until someone independant else submits the same results (his cd might have had no scratch, but is a different offset cd).

If you came along and ripped that cd and it had: 12345678 (confidence 7) then mathmatically the chances are:

4 Billion x 4 Billion x 4 Billion x 4 Billion x 4 Billion x 4 Billion x 4 Billion, or 4 Billion to the power of confidence. I would bet my life on anything with a confidence > 2 for sure.

It is a shame that there were cds with different pressing offsets because that lets in the possiblity of errors, for example 1 million people have the same CD, of these 1 million CDs half have scratches. In this instance (very extreme) those half million scratched cds that are submitted might have a few collisions on the CRC, this might appear that your cd (with a scratch is ripped accurate with confidence of 2), if there were not offset cds accuraterip would be able to report half a million people disagree with your results and only 2 do. There is something that can be done on the database back end to only create offset derrivative cds when the number is a percentage of submissions, so for 1 million cds you might need 50,000 matching to get that result in the database - this would counter that, but accuraterip does not have those sort of submission numbers where this is anywhere likely to be a problem.

If AccurateRip says something is accurate with a confidence of 3 or more, it is!

dammit. How do I help that then?



yes, i should have been more clear on my post - a lot of the cds i have are not logged on Accuraterip, and theres a very very high probability that they wont be, save for my entry. So ripping from 3 offset corrected drives and comparing CRCs and MD5s garentees that what I have is the best rip I can hope for for the cd at hand, not taking into account what others have



I dont doubt that! but if accuraterip doesnt have my cd, then its a little more difficult to try to garentee a perfect rip on my end.

I recommend you to read about offset and overreading in the coaster factory tutorial, the link is in the hydrogenaudio FAQs for EAC. Especially the offset pictures given there are almost self-explanatory.
Then try to find out your drives' capabilities.

I already corrected the offset of the drives thanks to accuraterip, so i'll look into the overread. thanks

And on a better note, you were right, precisionist. I asked around, and I got a reply from someone that has the one of the same discs I do, and "Keep your CD -- I hear a click at every single one of those spots on mine." sucks, to an extent, but im happy that its not a problem with the cd but with the shit pressing.

A little late now, but I'm surprised no one recommended trying an External ASPI layer such as those by Nero or Adaptec.

Also, EAC calculates track quality by the number of sectors to be read divided by the number of sectors it actually reads, except the initial 2 reads in secure mode count as 1. If you've got a drive that makes EAC re-read the last sector of each track you're likely to not get 100% unless the track is over 7m06s.

That´s because SPTI/ASPI either works or it dosen´t, but there will not be any differences in the extraction quality... -Martin.

You're right